Amid the prolonged Covid-19 pandemic, the discussion on how to distribute limited medical resources has emerged as a global topic. In the U.K. in particular, interest is growing in people with developmental disabilities placed in the blind spots of healthcare and welfare, although they account for only 1 percent of the total population. Reflecting the rising interest, a study has also begun to distribute medical resources to these people more efficiently.

A case in point is the project named “DECODE” (Data-driven machinE-learning aided stratification and management of multiple long-term COnditions in adults with intellectual DisabilitiEs). The project aims to establish a new integrated treatment model using artificial intelligence for people with developmental disabilities who suffer from complex diseases.

The U.K. government has high expectations about the project. DECODE has significance not only because it is a large-scale study that applies AI to the medical system but because development disorders and their parents are participating in it.

Accordingly, the National Health Service (NHS) and its research unit, the National Institute for Health Research (NIHR) have invested about 4.4 billion won (2.81 million British pounds) budget into the project.

The project is an investment in the future.

If medical professionals understand the interactions of these complex diseases, known as multiple long-term conditions (MLTCs), and predict their health effects and future diseases, it will help to improve the quality of life for people with learning disabilities by utilizing medical and welfare systems.

The project was launched in April last year under the initiative of Loughborough University and Leicestershire Partnership NHS Trust. Korea Biomedical Review met with Dr. Gyuchan Thomas Jun (Jun Gyu-chan), a Reader (Professor) in Socio-technical System Design at the School of Design and Creative Arts at Loughborough University, to hear about the meaning of this study and future plans. Professor Jun is a co-leader of the project.

Professor Jun is a design educator, system researcher and participatory design facilitator working in the fields of human-centered system design and system safety. He has been applying systems thinking approaches to healthcare system design and integrating new technologies, and teaching them, over the past two decades.

Question: The U.K. authorities have invested no less than 4.4 billion won into this project. Is there any background for this study to draw attention in the U.K.?

Answer: In the U.K., too, how to distribute medical resources fairly is an issue that attracts lots of attention. People with learning disabilities account for 1 percent of the total population in the U.K. Nevertheless, I was surprised to have an opportunity to conduct a study using such a large budget. It seems as if Covid-19 has increased interest in people with learning disorders, making it a political issue in the country and drawing attention.

Also, because the medical system is divided into a single disease treatment system, there is a huge investment in a single disease, but there was a lack of research or investment in complex diseases. Besides, there were limitations to understanding the patterns and clusters of MLTCs using existing statistical methods. The NHS also seemed to realize the importance of MLTCs and feel the need to invest in them. Moreover, the possibility of better understanding MLTCs based on AI, not the existing method, appeared to have moved the U.K. government toward making the ambitious investment.

Q: Why is it important to understand and predict the patterns and clusters of MLTCs?

A: Down syndrome causes problems in hearing, vision and heart. In this way, we can think that there's a pattern (of symptoms) when a cluster occurs. If we can know which development disability clusters cause epilepsy, kidney, hearing, and digestive problems, we can also know the disease’s progress. For instance, although it may not come out very accurately, but if someone shows epilepsy and developmental disabilities, he or she will have a 70 percent chance of a hearing problem in five years. Likewise, if you understand the cluster and progress of developmental disorders, you can respond to complex diseases accordingly when a new patient comes to you.

Also, people with learning disabilities tend to find it difficult to communicate their symptoms and pains, known as “diagnostic overshadowing” among medical professionals, which delays diagnosis and leads to untreatable situations. Therefore, one of this study’s purposes is to make predictions before symptoms occur. Family members also can make guesses when patients try to communicate, and make quick responses.

According to the NIHR, 67 percent of Britons with learning disabilities have two or more MLTCs, and they have 12 MLTCs on average.

Q: As it is a large-scale project with a large number of participants as well as a research budget, there may be many difficulties in conducting research.

A: To conduct this project, a team of 25 experts has been formed representing various walks of life, such as medical informatics, data science, AI, human engineering, system design, ethics, sociology, and psychiatry, who lead the entire project together with the specialists in developmental disabilities. We should feel grateful that we have data we can study, but we can't even get permission to access it. Each supplier of data has a different data-related policy. It takes more time than expected, but I think it will be solved eventually.

it is very important to let people with developmental disabilities participate in the study to think about whether the results from the data are really meaningful. We should create an environment and provide space to help them take part in the study comfortably. For instance, if there occurs a situation where we should hear only their parents’ opinions, we’ll need to prepare by, finding the patients’ carers and making researchers work out ways to provide cover. Accordingly, we included nurses and speech therapists in the team and are preparing to receive approval from the Institutional Review Board (IRB).

Q: What do you think are most crucial in this research?

A: The research team’s intention in carrying out this project is not to come up with certain insights from data but to take a holistic approach by taking out all changes that can occur from various aspects of new findings by allowing the participation of people with learning disabilities. When you submit a task, not only many experts but developmental disorder patients and their families examine it. I think it is significant because such a task does not end up as a floating story but a study to make meaningful changes. The two-and-a-half-year project also will not end with the current task but requires another study on a clinical decision support system (CDSS) based on AI. I hope the project will lead to a system that can predict diseases inflicting people with learning disabilities and an effective response system based on it.